Improvement of a FRET-based Indicator for cAMP by Linker Design and Stabilization of Donor–Acceptor Interaction Valentina Lissandron 1,2 , Anna Terrin 1,2 , Maddalena Collini 3 Laura D’alfonso 3 , Giuseppe Chirico 3 , Sergio Pantano 2 and Manuela Zaccolo 1,2 * 1 Dulbecco Telethon Institute at the Venetian Institute of Molecular Medicine, Via Orus 2 35129 Padova, Italy 2 Venetian Institute of Molecular Medicine, Via Orus 2, 35129 Padova, Italy 3 Laboratory of Advanced BioSpectroscopy, Physics Department, University of Milano-Bicocca, I 20126 Milano, Italy Fo ¨ rster resonance energy transfer (FRET) technology has been used to develop genetically encoded fluorescent indicators for a variety of intracellular molecular events. Often, however, the poor dynamic range of such reporters prevents detection of subtle but physiologically relevant signals. Here we present a strategy for improving FRET efficiency between donor and acceptor fluorophores in a green fluorescent protein (GFP)- based protein indicator for cAMP. Such indicator is based on protein kinase A (PKA) and was generated by fusion of CFP and YFP to the regulatory and catalytic subunits of PKA, respectively. Our approach to improve FRET efficiency was to perform molecular dynamic simulations and modelling studies of the linker peptide (L11) joining the CFP moiety and the regulatory subunit in order to define its structure and use this information to design an improved linker. We found that L11 contains the X-Y-P-Y-D motif, which adopts a turn-like conformation that is stiffly conserved along the simulation time. Based on this finding, we designed a new linker, L22 in which the YPY motif was doubled in order to generate a stiffer peptide and reduce the mobility of the chromophore within the protein complex, thus favouring CFP/YFP dipole–dipole interaction and improving FRET efficiency. Molecular dynamic simulations of L22 showed, unexpectedly, that the conformational behaviour of L22 was very loose. Based on the analysis of the three principal conformational states visited by L22 during the simulation time, we modified its sequence in order to increase its rigidity. The resulting linker L20 displayed lower flexibility and higher helical content than L22. When inserted in the cAMP indicator, L20 yielded a probe showing almost doubled FRET efficiency and a substantially improved dynamic range. q 2005 Elsevier Ltd. All rights reserved. Keywords: cAMP; protein kinase A; FRET; polypeptide conformation *Corresponding author Introduction In recent years, several genetically encoded fluorescent indicators have been developed based on the green fluorescent protein (GFP) and the phenomenon of Fo ¨ rster resonance energy transfer (FRET). 1 Such indicators provide a new means to dissect intracellular signal transduction by allowing direct visualization of individual molecular events as they take place in the intact cell. 2 The high resolution in space and time that such method- ologies provide opens the possibility to accurately measure quantitative and dynamic parameters of signalling networks in their complex cellular environment 3 and can provide novel insights for our understanding of biologically relevant pro- blems. 4,5 FRET relies on a non-radiative, distance-depen- dent transfer of energy from a donor chromophore to an acceptor chromophore by dipole–dipole 0022-2836/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. Abbreviations used: FRET, Fo ¨ rster resonance energy transfer; PKA, protein kinase A; GFP, green fluorescent protein; CFP, cyan fluorescentprotein; YFP, yellow fluorescent protein; C, catalytic; R, regulatory; MD, molecular dynamics. E-mail address of the corresponding author: manuela.zaccolo@unipd.it doi:10.1016/j.jmb.2005.09.089 J. Mol. Biol. (2005) 354, 546–555